> > Welcome to my world of m68k Macs. The ESP driver has no PDMA either,
> > right?
>
> I'm pretty sure that PDMA never worked on mac_esp, but it also wasn't
> as slow without it. The 5380 chip is pretty slow even compared to the
Well, it's definitely a lot slower than the 5380 with DMA.
> 53c94 and 53c96 in the models supported by mac_esp. Real DMA would be
> better, but only the IIfx and the AV macs even have the hardware.
>
> > Anyway - did the compiler both this, or did the code change?
>
> The pdma code is mostly asm, so I imagine something elsewhere in the
> code broke it. Someone mentioned the page fault handlers being
> involved, and I wouldn't be surprised at all if those changed.
That would have been me, perhaps. On second thought, I think the page
fault handler should not really get involved, the PDMA access part should
rather be protected by a page fault fixup section contained in the asm
code itself. I admit I haven't looked at the code in years, though. Just
guessing the compiler/assembler might mess up the fixup section now.
>From a quick glance: the exception table and fixup seems OK. We'd need a
register dump of the page fault to see more.
Michael
--
To UNSUBSCRIBE, email to [EMAIL PROTECTED]
with a subject of "unsubscribe". Trouble? Contact [EMAIL PROTECTED]
